Cuéntame: a local Gemma interviewer that preserves the words you actually said

Description and use

A local family-memory interview tool preserves the storyteller's exact words in a private Markdown file while Gemma only proposes questions; no fabricated biography is generated or stored.

Problem

Collect family memories through a conversational interview while keeping the storyteller's exact words verbatim, without letting an AI fabricate a biography.

Outcome

A local family-memory interview tool preserves the storyteller's exact words in a private Markdown file while Gemma only proposes questions; no fabricated biography is generated or stored.

Workflow

• Install Python 3.10+ and Ollama; pull the model gemma4:e2b via ollama pull gemma4:e2b
• Run python3 cuentame.py --nombre "Elena" --modelo gemma4:e2b to start the interview
• Gemma proposes each next question using earlier questions and answers
• The storyteller answers one short open question at a time
• Python copies the storyteller's exact answer verbatim into recuerdos/recuerdo-elena-ficticia.md
• The CLI refuses to overwrite existing memory files

Evidence

python3 cuentame.py --nombre 'Elena (ficticia)' --preguntas 1 ¿Cuál es un recuerdo de comida que te gustaría compartir sobre Elena? > Hacía arroz con coco los domingos. Guardado en recuerdos/recuerdo-elena-ficticia.md

Source

1. ¿Cuál es un recuerdo de comida que te gustaría compartir sobre Elena? > Hacía arroz con coco los domingos.

Source

The program rejects empty, multiline, overlong, or non-question replies.

Source

render_memory() writes the storyteller's responses as block quotes. It does not ask Gemma for a summary. That is a deliberate technical choice: a fluent summary can introduce a detail nobody said.

Source